Objectives: Memory and executive deficits are important cognitive markers of Alzheimer's disease (AD). Moreover, in the past decade, cerebrospinal fluid (CSF) biomarkers have been increasingly utilized in clinical practice. Both cognitive and CSF markers can be used to differentiate between AD patients and healthy seniors with high diagnostic accuracy. However, the extent to which performance on specific mnemonic or executive tasks enables reliable estimations of the concentrations of different CSF markers and their ratios remains unclear. Methods: To address the above issues, we examined the association between neuropsychological data and CSF biomarkers in 51 AD patients using hierarchical multiple regression analyses. In the first step of these analyses, age, education and sex were entered as predictors to control for possible confounding effects. In the second step, data from a neuropsychological test battery assessing episodic memory, semantic memory and executive functioning were included to determine whether these variables significantly increased (compared to step 1) the explained variance in A beta(42) concentration, p-tau concentration, t-tau concentration, A beta(42)/t-tau ratio, and A beta(42)/A beta(40) ratio. Results: The different models explained 52% of the variance in A beta(42)/t-tau ratio, 27% of the variance in A beta(42) concentration, and 28% of the variance in t-tau concentration. In particular, A beta(42)/t-tau ratio was associated with verbal recognition and code shifting, with A beta(42) being related to verbal recognition and t-tau being related to code shifting. By contrast, the inclusion of neuropsychological data did not allow reliable estimations of A beta(42)/A beta(40) ratio or p-tau concentration. Conclusion: Our results showed that strong associations exist between the cognitive key symptoms of AD and the concentrations and ratios of specific CSF markers. In addition, we revealed a specific combination of neuropsychological tests that may facilitate reliable estimations of CSF concentrations, thereby providing important diagnostic information for non-invasive early AD detection. (C) 2015 Elsevier Inc. All rights reserved.